Solid-State Nuclear Magnetic Resonance, Differential Scanning

24 Solid-State Nuclear Magnetic

Downloaded by UNIV OF SYDNEY on April 15, 2013 | http://pubs.acs.org Publication Date: May 5, 1990 | doi: 10.1021/ba-1990-0227.ch024

Resonance, Differential Scanning Calorimetric, and X-ray Diffraction Studies of Polymers Alan E. Tonelli, Marian A . Gomez , Hajime Tanaka , Frederic C. Schilling, Madeleine H . Cozine , Andrew J. Lovinger, and Frank A. Bovey 1

2

3

A T & T Bell Laboratories, Murray Hill, NJ 07974 High-resolution

NMR spectroscopy

was coupled with

differential

scanning calorimetry and X-ray diffraction techniques to study the solid-state structures,

conformations,

dynamics, and phase transi-

tions of several semicrystalline polymers. This combination of techniques was used to study the packing and dynamics of isotactic polypropylene chains in the α, ß, and smectic crystalline polymorphs; the conformations and dynamics of poly(diethyloxetane) and II crystals; the conformation

of poly(butylene

in

form

I

terephthalate)

chains in the α and strain-induced ß crystalline phases; the conformation and mobility of trans-1,4-polybutadiene

chains in the high

-temperature phase II crystals; the thermochromic phase transitions in several polydiacetylene

single-crystal and melt-crystallized

sam-

ples; and the thermotropic crystal to liquid-crystal transition in polyphosphazenes.

Comparison of the structures, conformations, and

dynamics of these polymer chains in their various solid phases provides a foundation upon which to build structure-property

relation-

ships.

'Current address: Institute de Ciencia y Tecnologia de Polimeros, C.S.I.C., Madrid, Spain *Current address: Department of Applied Physics, University of Tokyo, Tokyo, Japan Current address: Department of Chemistry, Yale University, New Haven, CT 06511 3

0065-2393/90/0227-0409$11.50/0 © 1990 American Chemical Society

In Polymer Characterization; Craver, C., et al.; Advances in Chemistry; American Chemical Society: Washington, DC, 1990.

410

h

POLYMER CHARACTERIZATION

IGH-RESOLUTTON, SOLID-STATE NMR SPECTROSCOPY was COUpled with

differential scanning calorimetry (DSC) and X-ray diffraction techniques to study the solid-state structures, conformations, dynamics, and phase transitions of several crystalline polymers. The techniques of cross-polarization (CP), high-power proton dipolar decoupling (DD), and rapid magic angle sample spinning (MAS) were applied at various temperatures to achieve high-resolution

1 3

C and

3 1

P N M R spectra from solid samples of several crys-

talline polymers. This chapter presents a review of that work. Results were obtained for poly(diethyloxetane) ( P D E O ) , £rans-l,4-polybutadiene (TPBD), isotactic polypropylene (i-PP), poly(butylene


eral

polydiacetylenes

(PDA),

and

terephthalate)

(PBT), sev-

poly(bis-4-ethylphenoxyphosphazene)

(PBEPP). Each of these polymers can be crystallized into two or more polymorphs. Observation of the chemical shifts and spin-lattice relaxation times, T j , for each of their chemically distinct nuclei permits an understanding of the conformations, packings, and mobilities of their chains in each of their solid phases monitored by D C S and X-ray diffraction measurements. The

1 3

C chemical shifts of polymers observed in high-resolution

1 3

C

N M R spectra of their solutions are sensitive to their microstructures, that is, stereoregularity, comonomer sequence, and defect structures (1). The microstructural sensitivity of polymer

l 3

C chemical shifts has its o r i g i n i n

the local polymer chain conformation (2, 3). Microstructural differences produce changes in the average local polymer chain conformation, which, in turn, are manifested as different

1 3

C chemical shifts for the carbon atoms in

the vicinity of each unique microstructure. The 7-gauche effect (2), as illustrated in Chart I, successfully accounts for the microstructurally dependent

1 3

C chemical shifts exhibited by poly-

mers in their high-resolution solution spectra. In addition, several examples from the high-resolution spectra of polymers in the solid state (4-10) indicate that the 7-gauche effect also importantly influences the

1 3

C chemical shifts

of solid polymers. The amorphous carbons in semicrystalline polyethylene (PE) resonate 2-3 ppm upheld from the crystalline carbons (6, 11, 12). This observation is expected because the crystalline carbons reside in the all-trans, planar zigzag conformation (no 7-gauche shielding), but the C - C bonds in the amorphous portions of P E possess some gauche character, and therefore the amorphous carbons experience 7-gauche shielding (Chart I). Bunn et al. (7) observed the methylene carbon resonance in crystalline syndiotactic polypropylene (s-PP) to be a doublet split by 8.7 ppm; for isotactic polypropylene (i-PP) the methylene carbon resonance is a singlet resonating midway between the s-PP methylene doublet. s-PP crystallizes (13) in the - T T G G -

conformation, in which half the methylene carbons

experience two 7-gauche effects a n d the remaining half experience n o 7gauche interactions. i-PP crystallizes (14) in the - T G T G - conformation, in which every methylene carbon experiences one 7-gauche shielding effect.


24.

T O N E L L I ET A L .

Solid-State NMR, DSC, and X-ray

Diffraction

411

(Q)


fi

0

P

NO y EFFECT ( C < 3. a - C H

should resonate at the same field in I and II, 2

II should resonate upfield from a - C H 1 by one 7(0) 2

or by 7(0) - 7(CH ), depending on whether bonds 3 and 4 3

are T or G in form II, and 4. C H (side chain) I should resonate upfield from C H (side chain) II by one 7(0) or by 7(0) - 7(CH ), depending on whether bonds 3 and 4 are T or G in form II (see Chart II). 2

2

3



24.

TONELLI

E T AL.

Solid-State NMR, DSC, and X-ray Diffraction

0

417

C H £

BOND 1 (T)

BOND 2 (T) (b)

sc

BOND 4 (T)

BOND 3 (T)

(C) Chart II. (a) PDEO chain structure, (b) Newman projections along the C-C backbone bonds (1,2) in PDEO. (c) Newman projections along the >C< CH side-chain bonds (3,4) in PDEO.

2

Table i n . Number of 7-gauche Interactions in Form I and II PDEO Carbon

Form I (T )

>C< CH a-CH

Form II

4

0 2 (CH ) 2

3

2 (CHa)

2

(T G ) 2

2

0 2 (CH ) 1(0) + if bonds 3 and 4 = T, 2 (CH ) if bonds 3 and 4 = G, 1 (CH ) 2

3

3

CH

2

(side chain)

2(0)

1(0) + if bonds 3 and 4 = T, 0 (CH ) if bonds 3 and 4 = G, 1 (CH ) 3

3


418

POLYMER

CHARACTERIZATION

A comparison of the three C P M A S spectra of P D E O ( M Figure 1 readily shows that the CH

n

=

50,000) in

C chemical shifts expected for > C
C
C C
C < and C H

3

carbons in P D E O . Thus, packing effects in P T O , P D M O , and P D E O can be as large as 2 ppm. However, packing effects of this magnitude are not nearly sufficient to explain the observed differences in the CH

1 3

C chemical shifts of the a - C H and 2

(side chain) carbons between forms I and II P D E O , which are

2

>4-6

ppm. In addition to interchain packing effects, some differences in intramolecular chain geometries must exist between forms I and II P D E O . Possibly the valence angles differ significantly between the T I crystals and the T G 2

4

conformation in form

conformation in the form II crystals. Such valence

2

angle differences would be expected (38) to produce large

C chemical shift

1 3

effects; however, it is a bit more difficult to understand why the > C < CH

and

carbons would not also be affected.

3

Suppose that the C - C - C a

a

backbone valence angle (6 ) is sensitive to aot

the rotational states of the C - C bonds (1 and 2 in Chart II) and adopts different values for the T and G rotational states. If 8 (G) > or < 8 (T), aa

then the valence angle C

s c

-C-C

s c

between ethyl side chains (0 ) should

follow (39), and 8 (T) > or < 8 (G). The SC

SC

aot

SC

1 3

C chemical shift of > C
b>4>) * glycol residue in the extended c f° structure (Chart III). However, the crystal structures proposed by Yokouchi et al. (4S) and Hall, Stambaugh, and co-workers (48-51) depart significantly from the extended, all-trans glycol structure. A l l crystal structures proposed for the relaxed, contracted a form approximate a gauche-trans-gauche conformation for the glycol residue, although there are differences in detail among them. a

r

n e

As noted by Davidson et al. (53), the low scattering power of hydrogen atoms makes X-ray diffraction unsuitable for defining the conformation of


420

POLYMER


CH

CH

2

CHARACTERIZATION

2

H

H

0

H

H

H

H

H

0

H

* = TRANS

^> = GAUCHE c

c

Chart III. PBT chain with torsions about C-C bonds indicated and Newman projections of the trans and gauche conformers about the terminal C-C bonds. the glycol residues in PBT. Instead, these same authors applied broad-line H N M R measurements to oriented P B T in both the relaxed a and strained 3 forms and determined the second moments of the proton line shapes as a function of specimen orientation. They found the *H N M R results to be consistent with a nearly fully extended (trans-trans-trans) conformation for the P stretched form and to agree quantitatively with the X-ray structure proposed by Hall and Pass (48). However, their N M R results for the relaxed a form were not consistent with any of the proposed crystal structures and suggest instead that the conformation and orientation of the central methylene pairs in the glycol residues are not substantially altered in the straininduced transformation from the a to p form. J

More recently, Grenier-Loustalot and Bocelli (54) studied the structures of four P B T model compounds by X-ray diffraction and high-resolution C N M R spectroscopy in the solid state. Single-crystal X-ray diffraction revealed that two of the P B T model compounds crystallized with trans-trans-transglycol residues; one had a trans-trans-gauche-glycol conformation, and the remaining compound had its glycol residue in the gauche-trans-gauche conformation. 1 3

In the high-resolution solid-state C N M R spectra of the P B T model compounds, they observed the central methylene carbons that are gauche to their ester oxygens to resonate 3.0-3.7 ppm upfield from those central methylene carbons adopting the trans arrangement (Chart III). This finding is consistent with the often observed shielding of carbon nuclei whose 7 substituents are in a gauche arrangement, that is, the 7-gauche effect (2, 3, 55). The central methylene carbons in a form P B T were observed to resonate midway between the corresponding methylene resonances in the model compounds. Because of the broadness of the central methylene resonance in P B T 1 3


24.

TONELLI E T AL.


421

(>3 ppm), which is likely a consequence of contributions from carbon nuclei in both the crystalline and amorphous regions of the sample (as discussed later), Grenier-Loustalot and Bocelli (54) were unable to draw conclusions regarding the conformation of the glycol residue in a-PBT. A similar study was attempted by Havens and Koenig (56), but it too was plagued by broad resonances and, in addition, by an erroneous conformational assignment to one of their P B T model compounds, as pointed out by Grenier-Loustalot and Bocelli (54). Most recently, Perry et al. (57) employed high-resolution solid-state C N M R techniques to study the crystalline conformations and dynamics of P B T chains in both the a and p crystalline forms. Their spectra also exhibited broad resonances, especially for the central methylene carbons (4-5 ppm). However, they concluded that, as the amount of trans content in the glycol residue increases, the interior methylene resonance shifts to a higher field, so the P-form resonance moves upfield from the a-form resonance. This conclusion is in direct opposition to the model compound study of GrenierLoustalot and Bocelli (54) and to the expected order of chemical shifts based on the conformationally sensitive 7-gauche effect (2, 3, 55).


1 3

In an attempt to determine the conformations of P B T chains in their aand P-form crystals, we conducted variable-temperature, high-resolution solid-state C N M R studies. Above ~100 °C, the spectra are significantly better resolved, with resonance line widths not exceeding 1-2 ppm. This narrowing of resonances is apparently a consequence (58) of removing contributions made by the amorphous carbons, which no longer cross-polarize efficiently at temperatures well above the glass transition of P B T (59). C o m parison of the high-temperature, high-resolution solid-state C N M R spectra recorded for a- and P-PBT with those of the P B T model compounds reported by Grenier-Loustalot and Bocelli (54) leads to several conclusions concerning the conformations of P B T chains in the relaxed a and stretched P crystals. 1 3

1 3

P B T pellets (Aldrich 19,094-2) were cryogenically ground to a fine powder, which was annealed at 150 °C for 3 days to produce the a form (45, 60). Melt pressing of the pellets at 250 °C produced thin films that were quenched into liquid nitrogen. Strips were cut from the film and placed in an Instron tensile testing machine. The strips were drawn to 300% elongation, held under tension, and annealed at 150 °C for several hours to produce (45) the P form sample. X-ray diffraction photographs recorded for both samples in the film form confirmed that we did indeed produce both a- and P-PBT samples. The P B T samples were spun in an aluminum oxide rotor with K e l - F [poly(chlorotrifluoroethylene)] end caps. P B T in the a form was placed in the rotor as a powder, while a strip of the P-form sample was wound under tension onto a spindle to form a spool, and the ends of the strip were glued to prevent relaxation to the a form. The spool wound with the P-form strip was then inserted into the rotor, and special end caps (Doty Scientific) were


422

POLYMER

CHARACTERIZATION

used to secure the ends of the spindle so that the (i-form P B T spool would rotate at the same speed as the rotor. A comparison of the C P M A S / D D spectra of a - and (3-PBT recorded at 105 °C is made in Figure 3. Table IV contains the observed solid-state C N M R chemical shifts. C spin-lattice relaxation times, T measured in the solid state at 105 °C, are given in Table V. 1 3


1 3

b

The comparison of C P M A S spectra recorded at 105 °C and presented in Figure 3 and the corresponding chemical shifts listed in Table IV show that, aside from the protonated aromatic carbons (PAR), the carbon nuclei in a - and (3-PBT resonate at nearly identical frequencies. This observation is at variance with the results of Perry et al. (57), who found the C P M A S / D D spectra of a - and 0 - P B T at 20 °C to be closely similar except for the

a

i|iiii|iiii|nii|iiii|iiiiniiniiii|iiinnn]inT|iin[iiii|ini|nii[iiii|rTn] 34

32

30

28

26

24

22

20

PPM

Figure 3. CPMAS/DD spectra measured at 105 °C for a- and 0 - P B T : full spectra (a) and expansion of the central methylene carbon regions (b).


24.


TONELLI E T AL.

423

Table IV. C Chemical Shifts of a- and P-PBT I3

Carbon C H OCH PAR NPAR 2

2

F E

c=o

a

g

Amorphous

27.2 66.2 130.8 135.2

27.6 66.7 131.7 135.2

26.6 65.9 130.2 134.9

165.6

165.5

166.0

0

NOTE: All C shifts are given in parts per million versus tetramethylsilane (TMS); spectra were measured at 105 °C and referenced to the POM resonance at 89.1 ppm from TMS (25), although we observed a 0.3-ppm downfield shift with respect to ambient temperature. "Measured from spectra obtained without CP, but with MAS/DD, using a 3-s pulse-repetition delay. 'NPAR stands for nonprotonated aromatic. 13


;

resonances of the interior methylene carbons. The source of this disparity is revealed by comparing the spectra in Figure 3 with those presented by Perry et al. (57). Perry's spectra are characterized by broad resonances, 2-3 times as broad as the resonances seen in Figure 3, presumably a consequence of the different local environments, both conformational and packing, experienced by the carbon nuclei of the sluggish, amorphous P B T chains, which result in a dispersion of chemical shifts. Recording the C P M A S / D D spectra of P B T at elevated temperatures (105 °C) results in enhanced resolution, because the amorphous carbons, which constituted 30-50% of the samples, are sufficiently above their glass transition temperature (T = 50-55 °C) (59) to be mobile enough not to cross-polarize efficiently (58). Comparison of the spectra recorded with and without C P (pulse-repetition delay of 3 s) makes apparent that the crystalline and amorphous carbons resonate at similar frequencies, and the amorphous carbon resonances appear upfield as expected. Thus at temperatures sufficiently close to T , where the amorphous P B T chains are relatively rigid and immobile, C P M A S / D D spectra would be expected (58) and in fact are observed to be significantly broadened by the overlap of crystalline and amorphous resonances. g

g

The near coincidence of methylene carbon chemical shifts observed for a - and p - P B T strongly suggests that in both the relaxed (a) and strained 0)

Table V. Spin-Lattice Relaxation Times, T for o> and p-PBT u

C H . OCH PAR NPAR C = Q 2

2

0.13 0.32 5.7 18.9 18.0

0.20 0.27 7.0 12.7 15.0

0.16 0.24 0.31 —

NOTE: All Ti values are in seconds and were measured at 105 ° C "Obtained under CP conditions with the Torchia (26) pulse sequence. 'Obtained without CP by the inversion-recovery method (27)


424

POLYMER

CHARACTERIZATION

crystals, the glycol residues of the P B T chains are adopting very similar conformations. Just what conformation is adopted by the glycol residues in crystalline PBT? Chart IV presents schematic structures of the four P B T model compounds studied by Grenier-Loustalot and Bocelli (54) and of PBT. A comparison of the central methylene carbon chemical shifts clearly indicates that the glycol residues in both a - and P - P B T crystals are in the nearly extended trans-trans-trans conformation found by Grenier-Loustalot and Bocelli (54) for the P B T model compounds 3 and 4. If the glycol residues of both a - and p - P B T are nearly fully extended,


then what conformational differences can account for the 10% increase in the fiber repeat of the P-form crystals that are formed upon extension of a PBT? The ester bonds [C( = 0 ) 0 ] in P B T are likely trans planar, as they are in the four P B T model compounds. In each crystalline model compound the 1 U

24.5 275

_

0 ) - c - o - c - c - c - c - o - c - ( 0 0

||

24.2

24.2

O ) - C - 0 - C - C r C - C - 0 - C - ( O

3 0

II 278 278 /-x 0)-c-o-c-c-c-c-o-c-(O) v

4 0

^

||

27.9 279

Ci - ^ O ) - C - 0 - C y C - C - j - C - 0 - C -

(Oj-C£

0 PBT

0 II

/~x

d-27.2 272 0-276 276

/-x

0 II

- o-c-(0)-c-o-c-c-c-c-o-c- (O)-c-oII

II

W

0

W

0

Chart IV. The four PBT model compounds studied by Grenier-Loustalot and Bocelli (54). The conformation of each glycol residue, as determined by X-ray diffraction, is indicated (t is trans; g is gauche), and the chemical shifts (parts per million versus tetramethylsilane) for the central methylene carbons observed by CPMAS/DD C NMR spectroscopy are also listed. The structure of PBT is presented, and the chemical shifts observed in this work for the central methylene carbons in the a- and $-form crystals are indicated. l3


24.

Solid-State N M R , D S C , and X-ray Diffraction

TONELLI E T AL.

435

C H - 0 bonds are also in the trans conformation. Because the chemical shifts 2

of the central methylene carbons should be sensitive (2, 3, 55) to the conformations of these bonds (see Charts III and IV), as well as to the conformations of the terminal C H - C H 2

2

bonds, the C H - 0 bonds in a- and 0 2

P B T are also most likely trans. If they were gauche in either or both polymorphs, then different chemical shifts would be expected for the central methylene carbons in a-and 0 - P B T , or chemical shifts reflecting this shielding (2) would be produced by 7-gauche carbonyl carbons. Instead, the chemical shifts of the central methylene carbons in both a- and 0 - P B T are very similar to the chemical shifts observed for the same carbons in model compounds 3 and 4, where both the C H - 0 and terminal C H - C H


2

2

2

bonds are

trans. The only conformational degree of freedom that remains to distinguish the a and 0 forms of P B T is the relative orientation of the carbonyl groups in the terephthaloyl residues, which are determined by the torsional angles about the s p - s p , carbonyl to aromatic C - C bonds (see Charts III and IV). 2

2

In terms of 7-gauche shielding effects (2, 3, 55), the chemical shifts of the protonated aromatic carbons (PAR) are expected to reflect the conformations of these bonds. In fact the 0.9-ppm chemical shift difference observed between the PARs of a- and 0 - P B T is by far the largest difference between their high-resolution, solid-state

1 3

C N M R spectra (see Figure 3). Thus our results support the

conclusion reached by Davidson et al. (53) via analysis of broad-line H N M R 1

data; changes in the conformation of the terephthaloyl residue, but not in the glycol residue, must accompany the solid-state transformation of P B T from the a to the 0 form. As suggested by the comparison of solid-state

1 3

C N M R chemical shifts

observed for a- and 0 - P B T and several of their model compounds, all of the bonds in crystalline P B T are nearly trans except the bonds connecting the ester groups to the aromatic rings. The bonds in amorphous P B T chains would be expected to be a mixture of trans and gauche conformations, the latter producing upfield chemical shifts via the 7-gauche effect. It follows that the amorphous carbon nuclei should resonate upfield from the corresponding carbons in a- and 0 - P B T . The solid-state

1 3

C N M R chemical shifts

presented in Table IV for the amorphous and crystalline carbons in P B T confirm this expectation. Spin-lattice relaxation times, T , for both crystalline forms and amorx

phous P B T , as presented in Table V, serve to indicate the motional characteristics of solid P B T chains. The most striking observation is the near coincidence of the T s measured for the methylene carbons of the glycol t

residues in the a and 0 crystallites of P B T with those observed for the amorphous methylene carbons. By contrast, the Tfi measured for the crystalline PAR carbons are more than an order of magnitude longer than those observed for the amorphous PAR carbons. Apparently, the methylene car-


426

POLYMER

CHARACTERIZATION

bons of the glycol residue are undergoing significant motion, independent of whether they are included in the a and 0 crystallites or not. In agreement with Perry et al. (57), whose measurements were performed more than 80 °C below ours (105 °C), we do not find any significant differences between the spin-lattice relaxation times of the crystalline methylene carbons in a - and 0 - P B T . Similarly, the T s of the crystalline carbon x

nuclei belonging to the terephthaloyl residues in a - and 0 - P B T are also not markedly different. The a and 0 phases apparently do not constrain the motions of their constituent P B T chains in any significantly different manner (also see Jelinski et al. (61) and Garbow and Schaefer (62) for further dis-


cussion of solid-state P B T motion). High-resolution, solid-state

1 3

C N M R studies of a - and 0 - P B T revealed

several important features concerning the conformations and motions of P B T chains in both crystalline phases. The glycol residues are in the nearly extended (trans-trans-trans) conformations in both crystalline forms, and different orientations of the ester groups and phenyl rings probably account for the 10% difference in the fiber repeats of a - and 0 - P B T . In both crystals the methylene carbons are sampling rapid motions, which are significantly faster than the motions experienced by the carbons of the terephthaloyl residues.

Isotactic Polypropylene (i-PP) (63) Isotactic polypropylene is a stereoregular vinyl polymer that normally develops significant crystallinity below 200 °C. The thermodynamieally stable crystalline form, or a-form, consists of i-PP chains in the 3

X

helical confor-

mation (.-.tgtgtg...) packed in a monoclinic unit cell (64, 65). Left- and righthanded helices are in close proximity. The metastable 0-form crystals of i-PP contain hexagonally packed 3 helical chains arranged in groups of the X

same helical handedness (left or right) resulting in the distant packing of left- and right-handed chains (65, 66). The smectic form of i-PP is (64, 67-73) only partially ordered compared to the a - and 0-crystalline forms, although the i-PP chains in the smectic form remain in the 3 helical conformation (64, 68). Smectic i-PP is therefore X

primarily disordered in the intermolecular packing of its chains. The purpose of the investigation reported here was to learn more about the structures of the crystalline regions in the a-, 0 - , and smectic forms of i-PP using high-resolution

1 3

C N M R spectroscopy as a structural probe.

Isotactic PP in the smectic form was made by cryogenic grinding of a Hercules Profax-6523 i-PP sample (74) as described by Lovinger et al. (60). The a-form i-PP was obtained from the smectic sample by annealing for 1 h at 160 °C. The 0-form sample was made (75) by unidirectional crystallization at a growth rate of 10 |xm/min with a temperature gradient of 300 °C/cm.



24.

TONELLI E T AL.


427

(c) smecllc

| i i 30

| i ? i i | i r i rj i i r t p i I r 25 20 15 10

i-pr-f-r—r-j > i i 1 p r r T 5 0 - 5 PPM

Figure 4. CPMASIDD spectra of i-PP in (a) a form, (b) 0/orm, and(c) smectic form. Spectra were recorded at ambient temperature with no reference employed. X - r a y diflractograms o f a l l three i - P P samples w e r e r e c o r d e d before a n d after the C N M R e x p e r i m e n t s to ensure that the h i g h - s p e e d (3 k H z ) m a g i c angle s p i n n i n g o f the samples d i d n o t i n d u c e a n y s o l i d - s o l i d transitions. 1 3

F i g u r e 4 presents t h e C P M A S / D D spectra o f the t h r e e forms o f i - P P r e c o r d e d at a m b i e n t t e m p e r a t u r e . T a b l e V I contains the o b s e r v e d solid-state C N M R c h e m i c a l shifts. C spin-lattice relaxation t i m e s , 2\, m e a s u r e d i n the s o l i d state at a m b i e n t t e m p e r a t u r e are g i v e n i n Table V I I . I n a d d i t i o n , the T values m e a s u r e d (76) for i - P P i n solution are also p r e s e n t e d i n T a b l e V I I for c o m p a r i s o n . A c o m p a r i s o n o f the C P M A S C N M R spectra i n F i g u r e 4 a n d t h e relative c h e m i c a l shifts p r e s e n t e d i n T a b l e V I leads to several observations. F i r s t , b o t h t h e m e t h y l e n e a n d m e t h y l c a r b o n resonances i n a - f o r m i - P P are split b y 1 p p m , as first r e p o r t e d b y B u n n e t a l . (77). T h e ratio o f intensities (peak heights) o f the d o w n f i e l d to the u p f i e l d c o m p o n e n t is 2:1 for b o t h c a r b o n types. B u n n et a l . (77) i n t e r p r e t e d this s p l i t t i n g as d u e to t h e i n e q u i v a l e n t sites, A a n d B , p r o d u c e d b y p a i r i n g o f helices o f opposite handedness (64) (see F i g u r e 5a), w h i c h are also p r e s e n t i n the ratio 1 3

1 3

v

1 3


428

POLYMER

Table VI.

CHARACTERIZATION

C NMR Chemical Shifts for i-PP in the Solid State

1 3

i-PP Form

CH

CH

CH

0, 1.07 0.37 0.47

0.20 0.02 0

0, 0.88 0.27 0.08

2

a

P

Smectic

3

N O T E : All C N M R chemical shifts are in parts per million and were observed at ambient temperature and referenced to the most upfield resonance of each carbon type. l 3

A:B

= 2:1. The A sites correspond to a separation of 5.28 A between helical


axes, and for the B sites the helices are 6.14 A apart. In the spectrum of 0-form i-PP (Figure 4b), on the other hand, each carbon exhibits a single resonance. The 0-form methylene and methyl resonances are close to the upfield member of each pair of the same resonances in the a form, which were attributed to the B sites (see Table VI). Figure 5b indicates the interchain packing proposed (65) for 0-form i-PP. Unlike the a-form packing (64), 3 i-PP helices of the same handedness are packed X

together in groups in the 0-form crystals. The interhelical separation (68) is 6.36 A , very similar to the smallest interhelical separation involving the B sites of a-form crystals. Thus, the near-coincidence of the

1 3

C chemical shifts

in 0-form i-PP with those corresponding to the B sites in the a form can be understood on the basis of similar interhelical separations in both packing modifications. Bunn et al. (77) also reported the i-PP.

1 3

C N M R C P M A S spectrum of 0-form

They found, in contrast to our results, that the chemical shifts of the

A site methylene and methyl carbon resonances in a-form i-PP were closer than the B site resonances to those observed in their 0-form i-PP. We were unable to obtain 0-form i-PP following their preparation method, a finding confirmed by X-ray diffraction. The difficulties in obtaining pure 0-form i PP exclusively via thermal treatment are noted in the literature (75, 78-81). Instead, special nucleating agents (80, 81) or unidirectional crystallization (75) are employed. Consequently, Bunn et al. (77) may not have studied i PP in the pure 0-form.

Table VII. i-PP Form a

P

Smectic Solution

1 3

C T Relaxation Times for Solid i-PP at Ambient Temperature

CH

CH

CH

37° 29 22 0.40*

52 34 33° 0.20*

0.32 , 0.80°, 0.48* 0.75°, 0.44* 0.75°, 0.51* 0.75*

a

fl

c

x

3

2

fl

fl

a

N O T E : All T i values are given in seconds. "Measured with CP using the pulse method of Torchia (26). ^Measured by inversion-recovery (27) without CP. 'Measured in solution at 46 °C by Randall (76) using the inversion-recovery method (27).



24.


TONELLI ET AL.

Diffraction

429

Figure 5. Crystal structures of (a) a form (64) and (b) $ form (65) of i-PP. Full (right-handed) and open (left-handed) triangles indicate 3j helical i-PP chains of different handedness. A and B label the inequivalent sites discussed in the text, and are applicable to all three carbon types, because the CH-CH bond is nearly parallel with the c-axis. Numerals at the triangle vertices indicate heights of methyl groups above a plane perpendicular to the c-axis in twelfths of c. The circles at the triangle vertices in part a correspond to methyl carbons, and the pairs of circles at numbers 10 and 4 correspond to the enmeshed Asite methyls. 2

1 3

C N M R chemical shifts recorded for the smectic form of i-PP are nearly

coincident with those found for the p form. This finding suggests that the local packing of 3

X

i-PP helices in the smectic form closely resembles that

found in the P form. Having concluded that the local packing of chains is similar in the P and smectic forms of i-PP on the basis of their observed

1 3

C chemical shifts, let

us look at the spin-lattice relaxation time (T ) behavior of the carbon nuclei t


430

POLYMER

in the three polymorphs of i-PP. The T

Y

CHARACTERIZATION

relaxation times of the crystalline

carbons were obtained while cross-polarizing by using the Torchia (26) pulse sequence. Smectic-form i-PP T

x

values (see Table VII) are very similar to

those measured for the P-form crystals, but the a-form T

x

values appear

unique. The T values reported previously for a-form i-PP by Fleming et l

al. (82) are in good agreement with the values shown in Table VII. The T values obtained by the inversion-recovery method (27) are domx

inated by the relaxation of the amorphous carbons and are obtained without CP. As expected from the spectra obtained without C P [not shown (63)], only the T of the methyl carbons are obtained by this method, and, within x


experimental error, they are the same for all three i-PP polymorphs. In addition, the T values measured for the methyl carbons in the crystals and t

in solution are similar to the amorphous methyl spin-lattice relaxation times. Clearly the spin-lattice relaxation times of methyl carbons are dominated by their internal rotations and not by the segmental motions of the i-PP chains. The two methyl resonances observed in the C P M A S spectrum of a-form i-PP (see Figure 4a) relax at different rates; T

{

= 0.32 and 0.80 s for the

downfield and upfield peaks, respectively. Having identified these resonances with the A and B packing sites in the a-form crystalline lattice (see Figure 5a), it is worth mentioning that both the (3- and smectic-form crystalline methyls have T

x

= 0.75 s in agreement with the T of the a-form, x

B-site methyl carbons. This observation supports the conclusion, obtained previously from a comparison of

1 3

C chemical shifts, that the interhelical

separation of chains is similar for the B sites in a-form crystals and in the 3- and smectic-form i-PP crystals. The a-form methyl carbons associated with the A and B packing sites exhibit T relaxation times different by almost a factor of 2. The B-site methyls Y

are rotating twice as fast as the enmeshed methyl carbons at the A sites because they are apparently on the fast side of the T minimum (82). x

The results of our solid-state

1 3

C N M R studies of the three polymorphs

of i-PP are consistent with several known structural features of the a - and P-form crystals and permit inferences about the local chain-packing structure in the smectic form. Both the

1 3

C chemical shifts and spin-lattice relaxation

times observed for the smectic-form carbons indicate that the packing of their 3! helices is similar (at least on a very local scale) to the packing of i PP chains in the p-form crystals. This conclusion is consistent with the proposal of Gailey and Ralston (69), who suggested that smectic-form i-PP is composed of small (50-100 A) hexagonal or P-form crystals. Suggestions that the smectic form is composed of monoclinic, or a-form, microcrystals made by Bodor et al. (70), or a smectic-form packing characterized locally by a core of a structure surrounded by chains in a pseudohexagonal arrangement made by Corradini et al. (71), are not consistent with the results of our

1 3

C N M R study. Also

the suggestion made by Miller (72) and Zannetti et al. (73) that a-form


24.

TONELLI ET AL.

Solid-State NMR,

DSC,

and X-ray

431

Diffraction

paracrystallinity (distortions of the monoclinic lattice with loss of long-range order) characterizes the structure of smectic-form i-PP is not supported by our results.

*rans-l,4-Polybutadiene (TPBD) (83) frans-l,4-Polybutadiene exists in two crystalline polymorphs (84). At room temperature, the chain conformation of form I is as follows (84, 85):


E

s

t

±

s*

E

• •• - C H - C H - C H = C H - C H - C H - C H = C H - C H - C H - - • • Z

2

2

2

where the double bond is of course trans ("E") and s

2

±

2

(or s ) T

designates

approximate skew conformations:

In the exact skew conformation, the C = C and C - H bonds are eclipsed, and the dihedral angle is 120°. The chain packing is a hexagonal array. Above approximately 75 °C, the stable form, called form II, is of lower density but with the chains still parallel to each other and still in a hexagonal array (86-89). They are believed to be in a disordered state, as judged by the blurring of all nonequatorial reflections in the X-ray diffraction pattern, and the marked decrease of the second moment of the wide-line proton N M R spectrum (90) indicates the onset of molecular motion. However, the details of the form II chain conformation and the nature of the motion are not well established. Suehiro and Takayanagi (86) proposed that the chain has a single definite structure, similar to that of form I except that the skew angle is decreased from 109° to 80°. They further proposed that the motion consists of large torsional oscillations about the carbon-carbon single bonds. Evans and Woodward (88) employed this conformation to calculate the heat capacity of form II and reported good agreement with experiment below and above the form I -> form II transition; they did not consider chain motion or attempt to calculate the heat capacity during the transition. Iwayanagi and Miura (90) proposed that, instead of undergoing large torsional oscillations,


432

POLYMER

CHARACTERIZATION


the chains are rotating about their long axes. Grebowicz and co-workers (92, 92) made thermodynamic calculations similar to those of Evans and Woodward (88) but assumed a conformationally disordered state for form II. DeRosa et al. (89) also proposed a disordered conformation—based on packing energy calculations—consisting of a 50:50 equilibrating mixture of a and b:

a:E

s (90°)

t

s*(90°)

b: E

s^O )

t

cis

±

0

E

E

This conformation corresponds to a 25% probability of cis for C H - C H bonds. We previously reported (93) the solid-state C N M R spectra of T P B D at room temperature. The results demonstrate that independent carbon nuclei can be observed from the crystalline and the mobile fold-surface regions of T P B D single crystals (94, 95). The olefinic and methylene carbons in the folds appear from their chemical shifts to have essentially the same average conformation as the 1,4-trans sequences in amorphous bulk polybutadiene. In addition, the C spin-lattice relaxation times (T ) of the folds are observed to be the same as for amorphous polymer (96), an observation indicating that chain motions in the two phases are similar. 2

1 3

1 3

x

Figure 6 shows C P M A S / D D spectra as a function of temperature. The spectrum at 23 °C, obtained by using a 1.0-ms contact time, shows single olefinic and methylene resonances for the crystalline stems of form I. As the temperature is increased, new resonances appear at higher field positions for both carbons, reflecting the onset of the solid-solid phase transition. At temperatures where both form I and II are present, a contact time of 2.0 ms was chosen to permit observation of both forms. However, because of substantial differences in chain mobility (vide infra), the intensities of these resonances do not quantitatively reflect the ratio of these phases. Despite this fact, we can estimate the midpoint of the transition to be ~60 °C; it is essentially complete at 65 °C. The midpoint observed in the initial heating of solution-crystallized T P B D is ~50 °C. The higher transition point observed following cooling and subsequent heating is probably the result of crystalline annealing during the first heating. At 23 °C the chemical shifts of the olefinic and methylene carbons of form I are 130.7 and 35.2 ppm, respectively (97). Those of form II are more shielded by 1.2 and 1.8 ppm, differences very close to those reported for fold-surface (i.e., amorphous) carbons versus crystalline stem carbons for form I (93). Despite the close similarity in chemical shift between the crystalline stem carbons of form II and amorphous carbons, individual resonances can be observed in non-cross-polarized spectra. These phases can be clearly differentiated in the course of an inversion-recovery (27) T measurement. x


24.


TONELLI E TAL.

o

b

b

Diffraction

433

a

-f CH -CH«CH-CH 45 2

2

E-b H-a

+65 °C


SSB

JL

SSB

Jy—s.

+62 °C

+23 «C SSB

1111 M i l l

200 Figure 6.

Tp

I-b

J

JL

IMIII111MIM11II | l 111M 11111IIII11111M1111II11111M11 M11M T I I I 11 | l

180 160 140 C NMR CPMAS/DD

13

SSB

120 100 80 ppm vs TMS

60

spectra, 50.31 MHz, of

40

IIIT f I M |

IIM111M

20

1,4-trms-polybuta-

diene (I is form 1; II is form II). SSB indicates spinning side bands.


434

POLYMER

CHARACTERIZATION

Figure 7 shows the 7\ spectra for the methylene carbons observed at 70 °C and therefore representing only form II. The fully relaxed spectrum ( T = 25 s) shows partial resolution of the two phases. Near the null, both positive and inverted resonances are observed, clearly indicative of two phases relaxing at different rates. As a result of the partial overlap of the methylene resonances in Figure 7, accurate T s cannot be determined from this inversion-recovery data. The T data measured by the cross-polarization method (CPT ) (26) are shown in Table VIII. The form II stem carbons exhibit values of 10.5 and 12.2 s for C H and C H , respectively. The value for the methylene carbons in the fold is estimated from the null point in Figure 7 to be ~0.7 s. These carbons are more shielded than the stem carbons by 0.6 ppm. For form I at 23 °C (Table VIII), the T values are 0.33 and 0.65 s for the surface-fold C H and C H , respectively, and the crystalline stem carbons exhibit much longer values having two components: 55 and 130 s for C H and 53 and 123 s for C H . The shorter values may correspond to monomer units near the crystal surface (98). x

x

X


2

x

2

2

The markedly greater shielding of the form II stem carbons is difficult to understand on the basis of the conformation proposed by Suehiro and Takayanagi (86), nor can the profound difference in carbon T values for form I and II stems be explained merely by torsional oscillations. Both observations seem consistent, however, with the disordered conformation suggested by DeRosa et al. (89), possibly combined with chain rotation, as proposed by Iwayanagi and Miura (90). x

Additional insight into the nature of the chain motion in T P B D can be obtained from observing the nonspinning C spectra as a function of temperature. The C P / D D spectrum at 23 °C in Figure 8a, corresponding to form I, shows that the olefinic carbon has an axially asymmetric chemical shift anisotropy, as previously reported (93). The value of a 11-^33 is ~178 ppm. The spectrum of form II recorded at 83 °C without C P (100) (Figure 8b) shows a dramatic narrowing in the powder pattern as a result of chain motion. The form of the pattern at 83 °C indicates a very anisotropic motion because the shift tensor does not simply average to the isotropic value, a ; instead the pattern of both the olefinic and methylene carbons shows an unsymmetrical change in addition to a growth in intensity at the isotropic positions. O f course, without C P a fraction of the intensity observed at a must be attributed to the amorphous carbons. 1 3

{

f

Additional evidence for the presence of conformational disorder is found in the differential scanning calorimetry (DSC) for T P B D , recorded on a Perkin-Elmer D S C - 4 with a heating rate of 10 °C/min. A large endotherm at 67 °C is associated with the form I —» form II transition, and a much smaller endotherm at 133.2 °C is associated with melting. The entropy of the solid-solid transition (AS ) in T P B D has been shown experimentally to be 1.6-2.0 times the entropy of melting (AS ), depending on sample preparation (101). tr

m



13

38

36

34

32

Figure 7. C NMR inversion-recovery spectra, 50.31 MHz, of the methylene carbons in 1,4-trzns-polybutadiene at 70 °C (II is form II; A is amorphous), T is inversion recovery delay time.

ppm vs TMS

iiiiiiii|uiiiiiniiiiiiMH|iiiiiiiii|iii n u n


436

POLYMER

Table VIII.

CHARACTERIZATION

C Cross-Polarization T Values for 1,4-frans-Polybutadiene

13

x

Stem Temp., ° C 23.0 50.5 60.0 60.0 70.0

Fold

Form

CH

CH =

I I I II II

55, 130 28, 69 23, 56 8.5 10.5

53, 123 40, 75 28, 66 9.1 12.2

2

CH

CH =

0.33°

0.65

2

— — —

~0.7

fo

a

— — — —


N O T E : All T\ values are given in seconds. "Inversion-recovery measurement. ^Estimate from inversion-recovery null point (Figure 7).

Earlier work (93, 99) showed that the treatment of T P B D crystals with m-chloroperbenzoic acid results in the epoxidation of the surface folds only and not the crystalline stems. The solid-state C N M R spectra showed that the epoxidation results in the immobilization of the surface folds and that the oxirane rings have probably raised the T of the fold surface above room temperature. Our present work supports this conclusion by the observation of a C T j of ~ 5 s for the oxirane carbons as compared to 0.6 s for the C H carbons in the folds of nonepoxidized T P B D . 1 3

g

1 3

We also examined the epoxidized polymer (28% total epoxidation) at various temperatures to observe the effect of this treatment upon the solid-solid transition. The sampled behavior changes little as compared to that of untreated T P B D . The C C P M A S / D D spectra shown in Figure 9 indicate that the midpoint of the transition is 47 °C, similar to the midpoint observed in the initial heating of the nonepoxidized T P B D material. Subsequent temperature cycling of the epoxidized sample does not change the midpoint of the transition because the immobilized surface prevents thickening of the crystals by annealing. In addition, examination of the X-ray diflraetograms of T P B D and epoxidized T P B D shows identical crystalline structures for the two materials as indicated by the position of the main reflection at 29 = 22.5°. The fact that the presence of immobilized oxirane rings on the surface of the crystal does not prevent or even perturb the solid-solid phase transition indicates that the folds are not involved in the phase transition of T P B D in any significant manner. Also, there is probably little motion along the direction of the crystalline stem in form II. Such motion would require movement of the oxirane folds, which appear to be immobile at 47 °C, as evidenced by the broadened resonances (Figure 9) for the oxirane C H and C H carbons. 1 3

2

High-energy irradiation of T P B D crystals initiates an expansion of the crystalline lattice at room temperature in a manner similar to that observed at the form I —» form II transition. Such irradiation will probably cause crosslinking, which in turn will inhibit motion within the crystalline regions of this analogue of form II.



24.

TONELLI ET AL.

i

i > i i

350

i i

Solid-State NMR,

i i

i i

i i

300

i i

i i i

250

DSC,

i

i i i

and X-ray

i i

200

i i

150

i i i

437

Diffraction

i i

100

i i

i i

i i i i i

50

0

8 Figure 8. C NMR nonspinning DD spectra, 50.31 MHz, of 1,4-trms-polybutadiene: (a) form I with CP, with a 5-s pulse-repetition delay, and (b)form II without CP, with a 60s pulse-repetition delay. 13

Polydiacetyknes (PDA) (102-105) P D A s p r e p a r e d (106) b y the t o p o c h e m i c a l p o l y m e r i z a t i o n o f single-crystal R

I

R

I

=c-oc-e= diacetlyene ( D A ) m o n o m e r s are a n i n t e r e s t i n g class o f p o l y m e r s , m u c h s t u d i e d (107, 108) because o f t h e i r availability as large single crystals a n d for t h e i r u n u s u a l o p t i c a l properties. P o l y ( E T C D ) , w h o s e substituent R is ( C H ) - O C O N H - C H - C H , is t y p i c a l o f the P D A s that e x h i b i t a t h e r 2

4

2

3

m o c h r o m i c t r a n s i t i o n , as e v i d e n c e d b y a change i n color from b l u e to r e d at ~ 115 °C i n the h e a t i n g process. (Also see F i g u r e 10.)


POLYMER

a b b a - t C H - C H • CH - C H 2

c rCH

2

d/\d c - C H - CH - C H

2

CHARACTERIZATION

2

FOLD

STEM

n-b

E-a


+62 °C

JL

SSB

SSB

d

JL

n-b

H-a

+53 °C I-a SSB

SSB

I-b

A

i

fl-a

n-b +47 *C

I-a

I-b SSB

SSB

JL I-a

I-b

+23 °C

SSB

x SSB

SSB

111111111111M1111M11f11111f t f1111fIT1111M111111111 f1111 It | f III M1111 m H 11MIU f 111111fIM111MI

200

180

160

140

120

100

80

60

40

20

ppm vs TMS ure 9. C NMR CPMAS/DD spectra, 50.31 MHz, of crystal surface epoxidized 1,4-trans-polybutadiene (I is form I; II is form II). I3


24.

TONELLI

E T AL.


Figure 11 shows the temperature dependence of

1 3

439

C N M R spectra in

the heating process. The coexistence of both phases can clearly be seen at 115 °C in the heating process. The

1 3

C chemical shifts and T s measured for x

poly(ETCD) are displayed for the blue (T < 115 °C) and red (T > 115 °C) crystalline phases in Table IX. We observed no resonances characteristic of the butatrienic form of backbone conjugation in either phase. The constancy R

R

I

I


- C = C = C = C of the C = 0 chemical shift implies that the hydrogen-bonded network of side chains (see poly(ETCD) structure) is retained in both phases. O n the basis of the differences in

1 3

C chemical shifts of the - C =

and P , 7 - C H

2

carbons in the blue and red phases, the thermoehromic blue-to-red phase transition observed in solid poly(ETCD) is most likely accompanied by a

c

\

c c /

C / N

VO'

/ 0

\ c

/

N

I

o \ c

/ c

c

\

\

I

/

c

\ \

c

\

//

/

\

\

/

c

/

c

c

\

c

/

c \ c

/

c

c

\

0

/

*\

c

c

\

/

0

c

1

\

N "

N

c

C

1 1

1 1

0

1

I c

\ c

The intramolecular structure of poly(ETCD) with hydrogen bonding. Protons are not drawn, and — represents the hydrogen bond.



440

POLYMER

400

140 T E M P E R A T U R E

CHARACTERIZATION

180 (°C)

Figure 10. DSC scans at 10 °C/min for (a) the first heating process, (b) the following cooling process, and (c) the second heating process to above the melting point in poly(ETCD).

planar-to-nonplanar conformational change in the backbone and an extension of the side chains to a nearly all-trans conformation. The chemical shift of the backbone - C = carbon is especially sensitive to the transition changing from 107 to 103 ppm on transition from the blue to the red phase. The observed chemical shift change of the resonance cannot be explained by defects in the backbone structure, because only a single resonance is observed for - C = in both phases. The backbone must be uniformly distorted. We have found (104) that, irrespective of side chains, all blue-phase P D A s have their - C = resonances at —107 ppm, and their


24.

TONELLI E T AL.


CH

2

CH 0-CH

441

C H

2

CH

3

2

2

(C)

' . C

=

"Cs


OO

I 111 i i i l l

160

i i lint i l l ii I 140 120

t l n m i i n Inn 100 80 PPM

Inn nm1 20 0

M i n i m u m i l i m n i ii

60

40

Figure 11. CPMAS/DD spectra of poly(ETCD) at 23 (a), 115 (b), and 127 (c) °C. Spectra were referenced to the resonance of POM (89.1 ppm from tetramethylsilane (TMS) (25)). The peaks labeled as SB correspond to the spinning side bands.


442

POLYMER

Table IX.

13

CHARACTERIZATION

C Chemical Shifts and Spin-Lattice Relaxation Times, Ti, for Poly(ETCD) R I

= c-c = c-c = I R R=CH -CH -- C H - C H - 0C0NH-CH -CH € Y 8 a 0 2

2

2

2

2

3

C (ppm vs. TMS)


13

Carbon C= 0 >C = -C = 5-CH a-CH €-CH

2

2

2

CH

3

-C=

Blue Phase (Low T) 157.5 131.6 107.4 66.6 37.3 32.9 24.5 16.2

Red Phase (High T)

Blue Phase (Low T)

Red Phase (HighT)

158.3 132,0 103.6 65.5 37.8 32.6 26.4 16.7

116 153 172 11, 110 9, 103 9, 108 2, 97 2, 14

47 25 30 6 8 6 4 6

carbons resonate at —103 ppm for red-phase P D A s . The backbone

structure of the blue phase was found to be more planar than that of the red phase on the basis of the overall - C =

shift positions. The transition

between blue and red phases is likely achieved by small rotations of opposite sign about the single C - C backbone bonds. The chemical shifts of the 0,7-

^csc4c

\

\

C H carbons are 2 ppm downfield in the red phase compared to their position in the blue phase. This result strongly suggests that the alkyl side-chain bonds have more trans or planar character in the red phase. Comparison of the chemical shifts of the 0,7 carbons in poly(ETCD) with those of model systems whose solid-state conformations are known (see Chart IV, for example), makes it appear that the alkyl portions of the side chains in poly(ETCD) have the gtg conformation in the blue phase and the t'tt' conformation in the red phase for the bonds of the tetramethylene fragment, where g,g are gauche rotations of opposite sign and t' represents an imperfect or nonplanar trans (t) conformation. This change in side-chain conformation is consistent with the expansion of the crystalline lattice in the side-chain 2


24.

TONELLI E T AL.


443

d i r e c t i o n o b s e r v e d b y X - r a y diffraction to accompany t h e transition from t h e b l u e to t h e r e d phase (105, 109). I n a d d i t i o n , t h e conformational transition gtg « ± t ' t t ' suggested for the a l k y l p o r t i o n o f the side chains is o f the H e l f a n d type (I JO) thought to b e the most facile for alkane chains i n c o n d e n s e d m e d i a . W h e n p o l y ( E T C D ) is r e c r y s t a l l i z e d , u p o n c o o l i n g from its m e l t , t h e - C = resonance moves to 102 p p m a n d t h e 0 , 7 - C H resonances to 2 7 . 5 p p m (105). T h u s , t h e b a c k b o n e appears e v e n m o r e nonplanar t h a n i n t h e r e d phase, a n d the side chains appear m o r e e x t e n d e d (ttt) w h e n c r y s t a l l i z e d from the m e l t ( F i g u r e 12).


2

0.728 nm

in

2.187 nm

1.096 nm 0.420 0.453 A n m

3Q

or

< H

in

z

100 200

300

UJ

H Z

u

a

UJ

H

o

< u. u.

100

200

UJ

302

JI

(b)

300 302 (a) J_

JL

J.

5 10 15 20 25 30 DIFFRACTION ANGLE 20 (°)

35

Figure 12. X-ray diffractograms of (a) single-crystal blue phase, (b) singlecrystal red phase, and(c) once-melted poly(ETCD). The 100 peak corresponds to the spacing between methyl carbons belonging to side chains attached to adjacent - C = C - carbons.


444

POLYMER

CHARACTERIZATION

Poly(bis-4-ethylphenoxyphosphazene) (PBEPP) (111, 112) Polyphosphazenes are an interesting class of inorganic polymers that can OR

I (-N=P- )


exhibit (113) several phase transitions dependent on their thermal histories. The most unique among these transitions is a thermotropic crystal-liquidcrystal transition, T(l), preceding the final melting. Figures 13 and 14 present D S C scans and X-ray diffractograms of P R E P P (R = C H C H C H ) that 6

reflect this transition. In Figure 15, the

3 1

4

2

3

P M A S / D D spectra of P R E P P

recorded at several temperatures are compared. The sudden decrease in line width observed above 100 °C (T(l)) reflects the crystal-liquid-crystal

0.1 c o l / g - K

I -30

i

I

i

10

l

i

50

TEMPERATURE

I

1—I

90 (°C)

Figure 13. DSC scans of PBEPP in the heating (a) and cooling (b) process.



24.

TONELLI E T AL.


I 2

I

I

I

I

I

I

6

10

14

18

22

26

DIFFRACTION A N G L E

445

1 30

20 ( ° )

Figure 14. X-ray diffractograms ofPBEPP at 25 (a) and 120 (b) °C. phase transition and indicates considerable backbone motion in the liquidcrystalline phase. This result is consistent with the disappearance of all intrachain diffraction peaks (see Figure 14) in the liquid-crystalline state (above 100 °C). The C P M A S / D D C N M R spectra of P B E P P are presented in Figure 16 both below and above the T(l) crystal-liquid-crystal transition. The spinlattice relaxation times, T , presented in Table X , indicate that the side chains are also mobile in the liquid-crystalline phase. In addition, the short T s observed only for the protonated aromatic carbons in the crystal indicate mobile phenyl rings rotating rapidly about their 1,4-axes even in this phase. However, crystalline phenyl-ring rotation is not rapid enough to average the chemical shifts of the methyl and several aromatic carbon resonances that are split into multiplets (Figure 16a). 1 3

x

Y

Conclusions We hope the several examples discussed here will serve to demonstrate the utility of high-resolution N M R spectroscopy to the study of solid polymers. Particularly when coupled with D S C and X-ray diffraction techniques, solid-


446



c

b

l 0

A

iiiiliiiiIiiiiIiiiiIiiiiIiiiiliiiiliiii | - 5

- 1 0

- 1 5

- 2 0

- 2 5

- 3 0

- 3 5

- 4 0

P P M

Figure 15. P MAS/DD spectra of PBEPP at 25 (a), 80 (b), and 120 (c) °C. 31

Table X.

Carbon C-a C-d C-b C-c CH CH

2

3

l3

C Spin-Lattice Relaxation Times, T PBEPP

b

for

T = 25 °C

T = 100 °C

17 15 1.5 1.5 10 2

4 3 0.6 0.5 0.8 2

NOTE: All T i values are given in seconds.


24.

T O N E L U

E T AL.


441


(b)

V

0^@VCH -CH 2

(a)

b

e

1

3

2

1

i1 a d

V

W

III|IIII|IIII|I1II|IHI|IIII|IIM|IIII|IIII |IIM|IIII|IIII1MII|IIII1IIII1

140

120 100 8 0

60

40

20

0

PPM Figure 16. C spectra below and above T(l): CPMAS/DD spectra at 24 (a) and 120 (b) °C. 13

state N M R spectroscopy can be a valuable tool for studying the structures, conformations, dynamics, and phase transitions of solid polymers, and will serve as a foundation upon which to build structure-property relations for polymers.

Acknowledgments We are grateful to M . Thakur for providing all P D A samples and to S. V. Chichester-Hicks and R. C . Haddon for synthesizing the sample of P B E P P .

American Chemical Society Library

15th St,Craver, N.W, C., et al.; In Polymer 1155 Characterization; Advances in Chemistry; American Chemical Society: Washington, DC, 1990.

448



References 1. Bovey, F. A. High-Resolution NMR of Macromolecules; Academic: New York, 1972. 2. Tonelli, A. E.; Schilling, F. C. Acc. Chem. Res. 1981, 14, 233. 3. Bovey, F. A. Chain Structure and Conformation of Macromolecules; Academic: New York, 1982. 4. VanderHart, D. L . ; Earl, W. L . ; Garroway, A. N. J. Magn. Reson. 1975, 44, 361. 5. Moller, M . Adv. Polym. Sci. 1985, 66, 59. 6. Earl, W. L . ; VanderHart, D. L. Macromolecules 1979, 12, 762. 7. Bunn, A.; Cudby, M . E . A.; Harris, R. K.; Packer, K. J.; Say, B. J. J. Chem. Soc. Chem. Commun. 1981, 15. 8. Belfiore, L. A.; Schilling, F. C.; Tonelli, A. E.; Lovinger, A. E.; Bovey, F. A. Macromolecules 1984, 17, 2561. 9. Sacchi, M . C.; Locatelli, P.; Zetta, L . ; Zambelli, A. Macromolecules 1983, 17, 483. 10. Terao, T.; Maeda, S.; Saika, A. Macromolecules 1983, 16, 1535 11. Kitamaru, R.; Horii, F.; Murayama, M . Polym. Bull. 1982, 7, 583. 12. Ando, I.; Yamanobe, T.; Sorita, T.; Komoto, T.; Sato, H . ; Deguchi, K.; Imanari, M. Macromolecules 1981, 17, 1955. 13. Corradini, P.; Natta, G.; Ganis, P.; Temussi, P. A. J. Polym. Sci. Part C 1967, 16, 2477. 14. Turner-Jones, A.; Aizlewood, J. M . ; Beckett, D. R. Makromol. Chem. 1961, 75, 134. 15. Natta, C . ; Corradini, P.; Bassi, I. W. Nuovo Cimento Suppl. 1960, 15, 52. 16. Miyashita, T.; Yokouchi, M . ; Chatani, Y.; Tadokoro, H . Annual Meeting of the Society of Polymer Science of Japan; Tokyo, 1974, preprint p 453. Quoted in Tadokoro, H . Structure of Crystalline Polymers; Wiley-Interscience: New York, 1979; p 405. 17. Turner-Jones, A. J. Polym. Sci. Part B 1963, 1, 455. 18. Petraccone, V.; Priozzi, B.; Frasci, A.; Corradini, P. Eur. Polym. J. 1976, 12, 323. 19. Zannetti, R.; Manaresi, P.; Buzzori, G. C. Chim. Ind. (Milan) 1961, 43, 735. 20. Danusso, F.; Gianotti, G. Makromol. Chem. 1963, 61, 139. 21. Geacintov, C.; Schottand, R.; Miles, R. B. J. Polym. Sci. Polym. Lett. Ed. 1963, 1, 587. 22. Miller, R. C.; Holland, V. F. J. Polym. Sci. Polym. Lett. Ed. 1961, 2, 519. 23. Schilling, F. C.; Bovey, F. A . ; Tonelli, A. E.; Tseng, S.; Woodward, A. E . Macromolecules 1981, 17, 728. 24. Perez, E.; VanderHart, D. L. Polymer 1987, 28, 733. 25. Earl, W. L . ; VanderHart, D. L. J. Magn. Reson. 1982, 48, 35. 26. Torchia, D. A. J. Magn. Reson. 1978, 30, 613. 27. Farrar, T. C.; Becker, E . D. Pulse Fourier Transform NMR; Academic Press: New York, 1971. 28. Gomez, M . A.; Cozine, M . H . ; Schilling, F. C.; Tonelli, A. E.; Bello, A.; Fatou, J. G. Macromolecules 1987, 20, 1761. 29. Perez, E.; Gomez, M . A . ; Bello, A.; Fatou, I. G. Colloid Polym. Sci. 1983, 261, 571. 30. Gomez, M . A.; Atkins, E . D. T.; Upstill, C.; Bello, A.; Fatou, J. G. Polymer, 1988, 29, 224. 31. Gomez, M . A.; Fatou, J. G.; Bello, A. Eur. Polym. J. 1986, 22, 43.


24.

TONELLI

ET

AL.

Solid-State

NMR,

DSC,

and

X-ray

449

Diffraction

32. Stothers, I. B. Carbon-13 NMR Spectroscopy; Academic: New York, 1972; Chapters 3 and 5. 33.

VanderHart, D . L . J. Magn. Reson. 1981, 44,

117.

34. VanderHart, D . L.; Khoury, F. Polymer 1983, 25, 1589. 35. Bunn, A.; Cudby, M. E . A . ; Harris, R. K . ; Packer, K . J.; Say, E . J. Polymer 1982, 23, 694. 36. c-(TO) actually adopts the (T2G2T2G2)2 conformation in the crystal (37) because of its cyclic structure. However, both the a and ß CH carbons experience the same kinds and numbers of 7-gauche interactions in both P T O and c-(TO) crystals. 4

2

4


37.

G r o t h , P. Acta Chem. Scand. 1971, 25,

725.

38. Reference 32, Chapter 4. 39. This is a consequence (40) of the conservation of total s and p character of the s p molecular orbitals around > C < . 3

40.

Huheey, J . E . Inorganic Chemistry: Principles of Structure and

Reactivity;

Harper and Row: N e w York, 1972; Chapter 4. 41. Gomez, M. A.; Cozine, M. H.; Tonelli, A. E . Macromolecules 1988, 21, 388. 42. Boye, C . A . , J r . ; Overton, J . R. Bull. Am. Phys. Soc. Sci. 2 1974, 19, 352. 43. Jakeways, R.; Ward, I. M.; Wilding, M. A.; H a l l , I. H.; Desborough, I. J.; Pass, M. G . J. Polym. Sci. Polym. Phys. Ed. 1975, 17,

799.

44. Jakeways, R.; Smith, T.; Ward, I. M.; Wilding, M. A . J. Polym. Sci. Polym. Lett. Ed. 1976,

14,

41.

45. Yokouchi, M.; Sakakibara, Y.; Chatani, Y.; Tadokoro, H.; Tanaka, T.; Yoda, K . Macromolecules

1976, 9,

266

46. Brereton, M. G.; Davies, C . R.; Jakeways, R.; Smith, T.; Ward, I. M. Polymer 1978, 19, 17. 47.

Mencik, Z. J. Polym. Sci. Polym. Phys. Ed. 1975, 13,

2173.

48. H a l l , I. H.; Pass, M. G . Polymer 1977, 17, 807. 49. Desborough, I. J.; H a l l , I. H. Polymer 1977, 18, 825. 50. Stambaugh, B. D . ; Koenig, J. L.; Lando, J. B. J. Polym. Sci. Polym. Lett. Ed. 1979,

15, 299; J. Polym. Sci. Polym. Phys. Ed. 1979, 171,

1053.

51. H a l l , I. H . In Fiber Diffraction Methods; French, A . D . ; Gardner, K . H., Eds.; A C S Symposium Series No. 141; American Chemical Society: Washington, D C , 1980, p 335. 52. Ward, I. M.; Wilding, M. A . Polymer 1977, 18, 327. 53. Davidson, I. S.; Manuel, A . I.; Ward, I. M. Polymer 1983, 24, 30. 54. Grenier-Loustalot, M . - F . ; Bocelli, G . Eur. Polym. J. 1983, 20, 957. 55.

Axelson, D . E . In High Resolution NMR

56. 57. 58. 59.

in Bulk; Komoroski, R. E., Ed.; VCH: Deerfield Beach, FL, 1986, Chapter 6. Havens, J. R.; Koenig, J. L. Polym. Commun. 1983, 24, 194. Perry, B. C . ; Koenig, J . L.; Lando, J . B. Macromolecules 1987, 20, 422. Lyerla, J. R.; Komoroski, R. A . ; Axelson, D . E . In ref. 55, Chapters 2-5. The T of P B T has been reported in the literature as varying from 22 to 80 °C

Spectroscopy of Synthetic Polymers

R

(see, for example: Lewis, O. G . Physical Constants of Linear

Homopolymers;

Springer-Verlag: New York, 1968; pp 150-151; and Lee, W. A . ; Knight, G . J. In Polymer Handbook; Brandrup, J.; Immergut, E . H., E d s . ; Interscience: N e w York, 1966; p III-79). D S C measurements performed on our a - P B T sample and a quenched amorphous P B T sample yielded T 50-55 °C. 60. Lovinger, A . J.; Belfiore, L . A . ; Bowmer, T. N . J. Polym. Sci. Polym. Phys. Ed. 1985, 23, 1449. 61. Jelinski, L. W.; Dumais, J . J.; Watnick, P. I.; Engel, A . K . ; Sefcik, M. D . g

Macromolecules

1983,

16,

409.

62. Garbow, J . R.; Schaefer, J. Macromolecules 1987, 20, 819.



450


63. Gomez, M . A.; Tanaka, H.; Tonelli, A. E . Polymer 1987, 28, 2227. 64. Natta, G . ; Corradini, P. Nuovo Cimento Suppl. 1960, 15, 40. 65. Turner-Jones, A.; Aizlewood, J. M . ; Beckett, D. R. Makromol. Chem. 1964, 75, 134. 66. Turner-Jones, A.; Cobbold, A. J. Polymer Lett. 1968, 6, 539. 67. Natta, G . ; Peraldo, M . ; Corradini, P. Rend. Accad. Naz. Lincei 1959, 26, 14. 68. Wyckoff, H . J. Polym. Sci. 1962, 62, 83. 69. Gailey, J. A . ; Ralston, P. H . Soc. Plast. Eng. Trans. 1964, 4, 29. 70. Bodor, G . ; Grell, M.; Kallo, A. Faserforsch. Textiltech. 1964, 15, 527. 71. Corradini, P.; Petraccone, V.; De Rosa, C.; Guerra, G. Macromolecules 1986, 19, 2689. 72. Miller, R. L. Polymer 1960, 1, 135. 73. Zannetti, R.; Celotti, G.; Armigliato, A. Eur. Polym. J. 1970, 6, 879. 74. C NMR of an i-PP solution (Profax) indicated that it contained 94% isotactic (mm) triads. 75. Lovinger, A. J . ; Ching, J. O.; Gryte, C. C. J. Polym. Sci. Polym. Phys. Ed. 1977, 15, 641. 76. Randall, J. C. J. Polym. Sci. Polym. Phys. Ed. 1976, 14, 1693. 77. Bunn, A.; Cudby, M . E . A.; Harris, R. K.; Packer, K. J . ; Say, E . J. Polymer 1982, 23, 694. 78. Varga, J.; Toth, F. Makromol. Chem. Macromol. Symp. 1986, 5, 213. 79. Shi, G . ; Huang, E.; Cao, Y.; He, Z.; Han, Z. Makromol. Chem. 1986, 187, 643. 80. Leugering, H . J. Makromol. Chem. 1967, 109, 204. 81. Duswalt, A. Am. Chem. Soc. Div. Org. Coat. 1970, 30, 93. 82. Fleming, W. W.; Fyfe, C. A.; Kendrick, R. D . ; Lyerla, J. R.; Vanni, H . ; Yannoni, C. S. In Polymer Characterization by ESR and NMR; Woodward, A. E . ; Bovey, F. A., Eds.; ACS Symposium Series No. 142, American Chemical Society: Washington, D C , 1980. 83. Schilling, F. C.; Gomez, M . A.; Tonelli, A. E.; Bovey, F. A.; Woodward, A. E . Macromolecules 1987, 20, 2954. 84. Natta, G . ; Corradini, P.; Porri, L. Atti Accad. Naz. Lincei Cl. Sci. Fis. Mat. Nat. Rend. 1956, 20, 728. 85. Natta, G.; Corradini, P. Nuovo Cimento Suppl. 1960, 1, 9. 86. Suehiro, J.; Takayanagi, M . J. Macromol. Sci. Phys. 1970, B4, 39. 87. Stellman, J. M . ; Woodward, A. E.; Stellman, S. D. Macromolecules 1973, 6, 330. 88. Evans, H.; Woodward, A. E . Macromolecules 1978, 11, 685. 89. DeRosa, C.; Napolitano, R.; Pirozzi, B. Polymer 1986, 26, 2039. 90. Iwayanagi, S.; Miura, I. Rep. Progr. Polym. Phys. 1965, 8, 1965. 91. Grebowicz, J . ; Lau, S.-F.; Wunderlich, B. J. Polym. Sci. Polym. Symp. 1984, 71, 19. 92. Grebowicz, J.; Aycock, W.; Wunderlich, E . Polymer 1986, 27, 575. 93. Schilling, F. C.; Bovey, F. A.; Tonelli, A. E.; Tseng, S.; Woodward, A. E . Macromolecules 1984, 17, 728. 94. Canale, A.; Hewett, W. A.; Shryne, T. M . ; Youngman, E . A. Chem. Ind. 1962, 1054. Rinehart, R. E . Polym. Prepr. (Am. Chem. Soc. Div. Polym. Chem.) 1966, 7, 556. 95. Wang, P.; Woodward, A. E . Macromolecules 1987, 20, 1818, 1823. 96. Jelinski, L. W.; Dumais, J. J . ; Watnick, P. I.; Bass, S. V.; Shepherd, L. I. Polym. Sci. Polym. Chem. Ed. 1982, 20, 3285. 97. All chemical shifts in this work are referenced to the amorphous methylene carbon at 32.79 ppm versus TMS (93). 1 3



24,

TONELLI ET AL.


451

98. Perez, E . ; VanderHart, D. L. J. Polym. Sci. Polym. Phys. Ed. 1987, 25, 1637. 99. Schilling, F. C.; Bovey, F. A.; Tseng, S.; Woodward, A. E . Macromolecules 1983, 16, 808. 100. The nonspinning spectrum recorded with CP at 83 °C shows a complete loss of intensity in the region of ov This distortion of the powder pattern for the mobile chains of form II is thought to result from a failure to cross-polarize those nuclei whose C - H vectors are oriented at or close to the magic angle of 54.7°. We thank Dr. L. W. Jelinski for bringing this point to our attention. 101. Ng, S. B.; Stellman, J. M . ; Woodward, A. E . J. Macromol. Sci. Phys. 1973, 7, 539. 102. Tanaka, H . ; Thakur, M . ; Gomez, M . A.; Tonelli, A. E . Macromolecules 1987, 20, 3084. 103. Tanaka, H . ; Gomez, M. A.; Tonelli, A. E.; Thakur, M . Macromolecules 1989, 22, 1208. 104. Tanaka, H . ; Thakur, M . ; Gomez, M . A.; Tonelli, A. E . unpublished data. 105. Tanaka, H . ; Gomez, M . A.; Tonelli, A. E.; Lovinger, A. J . ; Davis, D. D . ; Thakur, M . Macromolecules 1989, 22, 2427. 106. Wegner, G. Discuss. Faraday Soc. 1980, 68, 494. 107. Bloor, D.; Chance, R. R. Polydiacetylenes: NATO ASI Series E, Applied Science; Martinus Nijhoff Publisher (Kluwer Academic: Norwell, MA), 1985. 108. Chance, R. R. Encyclopedia of Polymer Science and Engineering; Wiley: New York, 1986; Vol. 4, p 767. 109. Downey, M . J . ; Hamill, G. P.; Rubner, M . ; Sandman, D. J.; Velazquez, C. S. Makromol. Chem. 1988, 189, 1199. 110. Helfand, E . J. Chem. Phys. 1971, 54, 4651. 111. Tanaka, H . ; Gomez, M. A.; Tonelli, A. E.; Chichester-Hicks, S. V.; Haddon, R. C. Macromolecules 1988, 21, 2301. 112. Tanaka, H . ; Gomez, M. A.; Tonelli, A. E.; Chichester-Hicks, S. V.; Haddon, R. C. Macromolecules 1989, 22, 1031. 113. Inorganic and Organometallic Polymers; ACS Symposium Series No. 360; American Chemical Society: Washington, D C , 1988, Chapters 19-25, pp 250-312. RECEIVED

30, 1989.

for review February 14, 1989.

ACCEPTED

revised manuscript November


Solid-State Nuclear Magnetic Resonance, Differential Scanning

Recommend Documents